Additive manufacturing device, additive manufacturing method, and machine learning device

By using laser beam flattening treatment during the unit weld bead formation process, the problem of residual gaps between weld beads in the directed energy deposition method is solved, and the strength and manufacturing quality of the product are improved.

CN115461185BActive Publication Date: 2025-10-03MITSUBISHI ELECTRIC CORP
View PDF 2 Cites 0 Cited by

Patent Information

Application Number
CN202080099939.1
Authority / Receiving Office
CN · China
Patent Type
Patents(China)
Current Assignee / Owner
Filing Date
2020-04-23
Publication Date
2025-10-03
Estimated Expiration
2040-04-23

AI Technical Summary

Technical Problem

In the prior art, when manufacturing three-dimensional structures using directed energy deposition, gaps easily remain between welds, resulting in reduced strength of the structure.

Method used

By using a laser beam flattening process when forming a unit weld bead, the unit weld bead is flattened, the contact angle is reduced and the formation of voids is suppressed. The laser beam irradiation is controlled by a control device to achieve the flattening of the unit weld bead.

Benefits of technology

It effectively suppresses the residual gaps between welds, improves the strength of the object, and ensures the high efficiency and high quality of the manufacturing process.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN115461185B_ABST
    Figure CN115461185B_ABST
Patent Text Reader

Abstract

An additive manufacturing device (100) manufactures a shaped object by stacking layers of solidified unit weld beads of molten material. The additive manufacturing device (100) comprises: a material supply unit (19) that supplies a material, i.e., a wire (5), to a workpiece; an irradiation unit that irradiates a laser beam (24) to melt the supplied material; and a control device (1) that forms unit weld beads by controlling the material supply unit (19) and the irradiation unit. When unit weld beads forming a layer are formed in contact with each other, the control device (1) flattens the formed unit weld beads by irradiating the beam, thereby forming a unit weld bead that contacts the flattened unit weld bead.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] The present invention relates to an additive manufacturing device, an additive manufacturing method and a machine learning device for manufacturing three-dimensional objects. Background Art

[0002] As one of the technologies for manufacturing three-dimensional objects, the technology of Additive Manufacturing (AM) is known. Among the multiple methods of additive manufacturing technology, the Directed Energy Deposition (DED) method has the advantages that the time taken to manufacture the object is short compared to other methods, and the material can be easily switched. The DED method has the advantage that there are fewer restrictions on the workpiece, that is, the base material, compared to other methods. In the case of the DED method, the consumption of material is limited to the amount used to manufacture the object, so there is less material waste compared to other methods. The additive manufacturing device of the DED method can appropriately change the structure of the processing head, thereby being able to use both powder and wire as materials. When using wire as the material, the welding wire of an existing product can be used, thereby suppressing the material procurement cost and making it easy to procure the material.

[0003] Patent Document 1 discloses a method for manufacturing a shaped object by stacking layers composed of multiple weld beads joined together. In the method disclosed in Patent Document 1, each of the multiple weld beads is formed by solidifying a weld line melted by an arc. Furthermore, in the method disclosed in Patent Document 1, once a layer is formed, the surface of that layer is melted before the next layer is formed, thereby suppressing any remaining gaps between the two stacked layers.

[0004] Patent Document 1: Japanese Patent Application Laid-Open No. 2019-63858 Summary of the Invention

[0005] According to the prior art disclosed in Patent Document 1, even if the surface of the layer is melted, gaps remain between adjacent weld beads in the same layer. Therefore, according to the prior art, gaps remain in the object, thereby reducing the strength of the object.

[0006] The present invention has been made in view of the above-mentioned circumstances, and an object thereof is to provide an additive manufacturing apparatus capable of suppressing a decrease in the strength of a formed object.

[0007] To address the aforementioned issues and achieve the objectives, the additive manufacturing apparatus of the present invention manufactures a shaped object by stacking layers of solidified unit weld beads (i.e., molten material) in parallel. The additive manufacturing apparatus of the present invention comprises a material supply unit that supplies material to the workpiece; an irradiation unit that irradiates a light beam to melt the supplied material; and a control device that forms the unit weld beads by controlling the material supply unit and the irradiation unit. As the unit weld beads, forming a layer, are formed in contact with each other, the control device flattens the formed unit weld beads by irradiating the unit weld beads with the light beam, thereby forming a unit weld bead that contacts the flattened unit weld beads.

[0008] Effects of the Invention

[0009] The additive manufacturing apparatus according to the present invention has the effect of suppressing a decrease in the strength of a formed object. BRIEF DESCRIPTION OF THE DRAWINGS

[0010] Figure 1 This is a diagram showing an additive manufacturing device according to the first embodiment.

[0011] Figure 2 This is a schematic diagram showing a state of processing by the additive manufacturing apparatus according to the first embodiment.

[0012] Figure 3 This is a block diagram showing a hardware configuration example of a control device included in the additive manufacturing apparatus according to the first embodiment.

[0013] Figure 4 This is a flowchart showing the operation procedure of the additive manufacturing apparatus according to the first embodiment.

[0014] Figure 5 This is a diagram for explaining the formation of a deposit by the additive manufacturing apparatus according to the first embodiment.

[0015] Figure 6 This is a schematic diagram showing a unit weld bead formed by the additive manufacturing apparatus according to the first embodiment.

[0016] Figure 7 This is a diagram for explaining a comparative example of the first embodiment.

[0017] Figure 8 This is a flowchart showing the operation procedure of the additive manufacturing apparatus according to the second embodiment.

[0018] Figure 9 This is a diagram for explaining the formation of a deposit by the additive manufacturing apparatus according to the second embodiment.

[0019] Figure 10This is a flowchart showing the operation procedure of the additive manufacturing apparatus according to the third embodiment.

[0020] Figure 11 This is a diagram for explaining the formation of a deposit by the additive manufacturing apparatus according to the third embodiment.

[0021] Figure 12 This is a flowchart illustrating a method for forming a beaded weld by the additive manufacturing apparatus according to the third embodiment.

[0022] Figure 13 This is a diagram for explaining the formation of a beaded weld by the additive manufacturing apparatus according to the third embodiment.

[0023] Figure 14 1 is a schematic plan view of a beaded weld formed by the additive manufacturing apparatus according to the third embodiment.

[0024] Figure 15 This is a diagram showing the configuration of an additive manufacturing system according to a fourth embodiment.

[0025] Figure 16 This is a flowchart showing the operation procedure of the machine learning device involved in embodiment 4. DETAILED DESCRIPTION

[0026] Hereinafter, an additive manufacturing device, an additive manufacturing method, and a machine learning device according to the embodiments will be described in detail with reference to the accompanying drawings.

[0027] Implementation method 1.

[0028] Figure 1 This is a diagram showing an additive manufacturing device according to the first embodiment. Figure 2 This is a schematic diagram illustrating processing performed by the additive manufacturing apparatus according to Embodiment 1. Additive manufacturing apparatus 100 is a machine tool that produces a shaped object by adding molten material to the workpiece. In Embodiment 1, the light beam is a laser beam 24, and the material is a metal wire 5. Wire 5 may also be a material other than metal. The material used in additive manufacturing apparatus 100 is not limited to wire 5; metal or resin powder may also be used.

[0029] Additive manufacturing apparatus 100 manufactures a shaped object by stacking layers of solidified molten material, or unit weld beads, to form a deposit 18. The shaped object is the deposit 18 obtained after the addition of material according to the processing procedure is completed. Additive manufacturing apparatus 100 forms deposit 18 on base material 17. Base material 17 is placed on a worktable 15. Figure 1The base material 17 shown is a plate material. The base material 17 may also be a material other than a plate material. In the following description, the workpiece refers to an object to which the molten material is attached, and refers to the base material 17 or the deposit 18.

[0030] The additive manufacturing apparatus 100 includes a processing head 10 that moves relative to a workpiece. The processing head 10 includes a beam nozzle 11, a wire supply nozzle 12, and a gas nozzle 13. The beam nozzle 11 emits a laser beam 24 toward the workpiece. The laser beam 24 is a heat source that melts the wire 5. The wire supply nozzle 12 moves the wire 5 toward the irradiation position of the laser beam 24 in the workpiece. The gas nozzle 13 sprays an inert gas 25 as a shielding gas toward the workpiece. The additive manufacturing apparatus 100 suppresses oxidation of the deposit 18 by spraying gas and cools the layer formed on the workpiece. The beam nozzle 11, the wire supply nozzle 12, and the gas nozzle 13 are fixed to the processing head 10, thereby uniquely determining their relative positions. In other words, the relative positions of the beam nozzle 11, the gas nozzle 13, and the wire supply nozzle 12 are fixed.

[0031] The laser oscillator 2, serving as a beam source, oscillates and generates a laser beam 24. The laser beam 24 from the laser oscillator 2 is transmitted to the processing head 10 via an optical cable 3, which is an optical transmission path. The laser oscillator 2, the optical cable 3, and the processing head 10 constitute an irradiation unit that irradiates the workpiece with the laser beam 24, which melts the wire 5.

[0032] The laser beam 24 irradiated from the beam nozzle 11 toward the workpiece and the center axis CW of the line 5 can be arranged non-coaxially or coaxially. By using a ring-shaped annular beam as the laser beam 24 and thereby using a laser beam branched into multiple beams as the laser beam 24, the laser beam 24 irradiated from the beam nozzle 11 toward the workpiece and the center axis CW of the line 5 can be arranged coaxially. Furthermore, in the first embodiment, the case where the laser beam 24 irradiated from the beam nozzle 11 toward the workpiece and the center axis CW of the line 5 are non-coaxial is described.

[0033] The gas supply device 7 supplies the inert gas 25 to the gas nozzle 13 via the pipe 8. The gas supply device 7, the pipe 8, and the gas nozzle 13 constitute a gas supply unit that ejects the inert gas 25 toward the processing area 26.

[0034] The wire reel 6, around which the wire 5 is wound, serves as the material supply source. The wire reel 6 rotates in response to the drive of the servo motor, or rotary motor 4, thereby unwinding the wire 5 from the reel 6. The wire 5 unwinding from the reel 6 is supplied to the laser beam 24 irradiation position through the wire supply nozzle 12. Furthermore, by rotating the rotary motor 4 in the opposite direction to that in which the wire 5 is unwinding from the reel 6, the wire 5 supplied to the laser beam 24 irradiation position can be pulled out from the laser beam 24 irradiation position. In this case, a portion of the wire 5 unwinding from the reel 6 on the reel 6 side is wound onto the reel 6. The rotary motor 4, the reel 6, and the wire supply nozzle 12 constitute a material supply unit 19 that supplies material to the workpiece.

[0035] Furthermore, the wire supply nozzle 12 may be provided with an operating mechanism for pulling the wire 5 out from the wire reel 6. The additive manufacturing apparatus 100 is provided with at least one of the rotating motor 4 and the operating mechanism of the wire supply nozzle 12, thereby being able to supply the wire 5 to the irradiation position of the laser beam 24. Figure 1 In the figure, the operating mechanism of the thread supply nozzle 12 is omitted.

[0036] The processing head driving device 14 moves the processing head 10 in each of the X-axis direction, the Y-axis direction and the Z-axis direction. The X-axis, the Y-axis and the Z-axis are three axes perpendicular to each other. The X-axis and the Y-axis are axes parallel to the horizontal direction. The Z-axis direction is the vertical direction. The processing head driving device 14 has a servo motor constituting an action mechanism for moving the processing head 10 in the X-axis direction, a servo motor constituting an action mechanism for moving the processing head 10 in the Y-axis direction, and a servo motor constituting an action mechanism for moving the processing head 10 in the Z-axis direction. The processing head driving device 14 is an action mechanism capable of performing translational motion in each of the three axes. Figure 1 The servo motors are not shown in the figure. The additive manufacturing apparatus 100 moves the machining head 10 via the machining head drive 14, thereby moving the irradiation position of the laser beam 24 on the workpiece. The additive manufacturing apparatus 100 moves the worktable 15, thereby moving the irradiation position of the laser beam 24 on the workpiece.

[0037] exist Figure 1 In the machining head 10 shown, the laser beam 24 is directed from the beam nozzle 11 in the Z-axis direction. The wire supply nozzle 12 is positioned in the XY plane at a distance from the beam nozzle 11 and advances the wire 5 in a direction inclined relative to the Z-axis. Furthermore, the wire supply nozzle 12 is fixed in the machining head 10 in the Z-axis direction, allowing the wire 5 to be advanced parallel to the Z-axis. The wire supply nozzle 12 regulates the advancement of the wire 5 so that the wire 5 is supplied to the desired position.

[0038] exist Figure 1In the machining head 10 shown, the gas nozzle 13 is positioned coaxially with the beam nozzle 11 in the XY plane, on the outer periphery of the beam nozzle 11. The gas nozzle 13 ejects gas along the central axis of the laser beam 24 emitted from the beam nozzle 11. In other words, the beam nozzle 11 and the gas nozzle 13 are coaxially positioned. Furthermore, the gas nozzle 13 can eject gas in a direction tilted relative to the Z axis. That is, the gas nozzle 13 can eject gas in a direction tilted relative to the central axis of the laser beam 24 emitted from the beam nozzle 11.

[0039] The rotating mechanism 16 is an operating mechanism capable of rotating the table 15 about a first axis and a second axis perpendicular to the first axis. Figure 1 In the rotating mechanism 16 shown, the first axis is an axis parallel to the X axis, and the second axis is an axis parallel to the Y axis. The rotating mechanism 16 has a servo motor constituting an action mechanism for rotating the worktable 15 around the first axis, and a servo motor constituting an action mechanism for rotating the worktable 15 around the second axis. The rotating mechanism 16 is an action mechanism capable of performing rotational motion around each of the two axes. Figure 1 The illustration of each servo motor is omitted.

[0040] The additive manufacturing apparatus 100 can change the posture or position of the workpiece by rotating the worktable 15 using the rotation mechanism 16. Specifically, the additive manufacturing apparatus 100 can move the irradiation position of the laser beam 24 within the workpiece by rotating the worktable 15. The use of the rotation mechanism 16 also enables the formation of complex shapes, including tapered ones.

[0041] The control device 1 controls the additive manufacturing apparatus 100 according to the machining program. The control device 1 is, for example, a numerical control device. The control device 1 outputs position commands to the machining head drive 14, thereby controlling the position of the machining head drive 14. The control device 1 outputs commands (indications corresponding to beam intensity conditions) to the laser oscillator 2, thereby controlling the laser oscillation of the laser oscillator 2.

[0042] The control device 1 outputs a supply command, or command corresponding to the material supply amount condition, to the rotary motor 4, thereby controlling the rotary motor 4. The supply command may be a command corresponding to the supply speed condition of the wire 5. The supply speed is the speed of the wire 5 moving from the wire reel 6 toward the irradiation position. The supply speed indicates the amount of material supplied per time.

[0043] The control device 1 outputs a command corresponding to the gas supply amount conditions to the gas supply device 7, thereby controlling the amount of inert gas 25 supplied from the gas supply device 7 to the gas nozzle 13. The control device 1 outputs a rotation command to the rotation mechanism 16, thereby controlling the drive of the rotation mechanism 16. In other words, the control device 1 outputs various commands to control the entire additive manufacturing apparatus 100. The control device 1 forms a unit weld bead in the additive manufacturing apparatus 100 by controlling the material supply unit 19, the irradiation unit, the gas supply unit, the processing head drive unit 14, and the rotation mechanism 16.

[0044] The additive manufacturing device 100 is as follows Figure 2 As shown, by irradiating the wire 5 supplied to the processing region 26 with a laser beam 24, molten material 21 is deposited in the processing region 26. In the processing region 26, the workpiece is melted at the surface 22 of the workpiece to form a molten pool 23. In the processing region 26, the molten material 21 generated by the melting of the wire 5 is welded to the molten pool 23. The processing region 26 is an area on the surface 22 where additional processing is performed.

[0045] The additive manufacturing apparatus 100 moves the machining head 10 and the table 15 by interlocking the machining head drive 14 and the rotation mechanism 16, thereby changing the position of the machining area 26 on the surface 22. Thus, the additive manufacturing apparatus 100 can obtain a desired shape of the product.

[0046] Next, a description will be given of the hardware configuration of the control device 1. The functions of the control device 1 are realized by executing a control program for controlling the additive manufacturing apparatus 100 using hardware.

[0047] Figure 3 This is a block diagram showing an example of the hardware configuration of a control device included in the additive manufacturing apparatus according to Embodiment 1. The control device 1 includes a CPU (Central Processing Unit) 41 that executes various processes, a RAM (Random Access Memory) 42 containing a data storage area, a ROM (Read Only Memory) 43 that is a nonvolatile memory, a storage device 44, and an input / output interface 45 for inputting and outputting information to and from the control device 1. Figure 3 The various components shown are connected to each other via a bus 46 .

[0048] The CPU 41 executes a program stored in the ROM 43 or the storage device 44. The control of the entire additive manufacturing apparatus 100 by the control device 1 is realized using the CPU 41.

[0049] The storage device 44 is an HDD (Hard Disk Drive) or SSD (Solid State Drive). The storage device 44 stores control programs and various data. The ROM 43 stores programs for basic control of the control device 1 (i.e., a computer or controller), such as a boot loader like the BIOS (Basic Input / Output System) or UEFI (Unified Extensible Firmware Interface), and software or programs for controlling the hardware. The control program may also be stored in the ROM 43.

[0050] Programs stored in ROM 43 and storage device 44 are downloaded to RAM 42. CPU 41 expands the control program in RAM 42 and executes various processes. Input / output interface 45 is an interface for connecting external devices to control device 1. Processing programs are input to input / output interface 45. In addition, input / output interface 45 outputs various commands. Control device 1 may include input devices such as a keyboard and pointing device, and output devices such as a display.

[0051] The control program can be stored in a computer-readable storage medium. The control device 1 can store the control program stored in the storage medium in the storage device 44. The storage medium can be a floppy disk (i.e., a removable storage medium) or a semiconductor memory (i.e., a flash memory). The control program can be installed from another computer or server device to the computer or controller serving as the control device 1 via a communication network.

[0052] The functions of the control device 1 can be implemented by dedicated hardware, namely, a processing circuit, for controlling the additive manufacturing apparatus 100. The processing circuit can be a single circuit, a complex circuit, a programmed processor, a parallel programmed processor, an ASIC (Application Specific Integrated Circuit), an FPGA (Field-Programmable Gate Array), or a combination thereof. The functions of the control device 1 can be partially implemented by dedicated hardware and partially implemented by software or firmware.

[0053] Next, refer to Figure 4 and Figure 5 , the operation of the additive manufacturing apparatus 100 according to the first embodiment will be described. Figure 4 This is a flowchart showing the operation procedure of the additive manufacturing apparatus according to the first embodiment. Figure 5 This is a diagram for explaining the formation of a deposit by the additive manufacturing apparatus according to the first embodiment.

[0054] The additive manufacturing apparatus 100 manufactures an object by stacking layers composed of multiple parallel unit weld beads. A unit weld bead is a solidified product formed by a single processing operation. A processing operation begins by adding molten material 21, continues fusing the molten material 21 to the molten pool 23, and then stops adding molten material 21. At the end of each processing operation, the additive manufacturing apparatus 100 moves the processing head 10 to the position where the next processing operation will begin. In the first embodiment, the additive manufacturing apparatus 100 adds molten material 21 while moving the processing head 10 in a linear direction, thereby forming linear unit weld beads.

[0055] In step S1, the additive manufacturing apparatus 100 forms a unit weld bead 51a on the workpiece, i.e., the base material 17. In step S2, the additive manufacturing apparatus 100 irradiates the edge 52 of the formed unit weld bead 51a with a laser beam 24. The edge 52 is the edge on the side where the next unit weld bead is to be formed, of the two edges along the length of the unit weld bead 51a. The additive manufacturing apparatus 100 scans the edge 52 with the laser beam 24, thereby irradiating the entire edge 52 with the laser beam 24. The additive manufacturing apparatus 100 flattens the unit weld bead 51a by irradiating the edge 52 with the laser beam 24. In the following description, the process of flattening the formed unit weld bead may be referred to as a flattening process. In the first embodiment, the flattening process forms a flat portion 53 on the edge 52 side of the unit weld bead 51a.

[0056] Figure 6 Schematic diagram showing a unit weld bead formed by the additive manufacturing apparatus according to the first embodiment. Figure 6 One longitudinal end face of a unit weld bead 51a is shown. Unit weld bead 51a is a three-dimensional object with a width Wx in the X-axis direction and a height H in the Z-axis direction. Height H is the height between the workpiece surface 22 and the top 55. In the following description, the ratio of width Wx in the X-axis direction to height H, or Wx / H, is sometimes referred to as flatness. Top 55 is the point in unit weld bead 51a where the vertical height relative to surface 22 is highest.

[0057] The edge 52 represents the three contact points between the unit weld bead 51a, the workpiece surface 22, and the gas surrounding the unit weld bead 51a. The contact angle θa is the angle between the surface 22 and the surface of the unit weld bead 51a, the angle including the unit weld bead 51a. The contact angle θa decreases after the flattening process compared to before the flattening process. The additive manufacturing apparatus 100 irradiates the laser beam 24 by shifting the irradiation position of the laser beam 24 in the X-axis direction from the irradiation position during formation of the unit weld bead 51a, thereby aligning the central axis of the laser beam 24 with the edge 52.

[0058] During the flattening process, the central axis of the laser beam 24 may be offset from the edge 52. The offset between the edge 52 and the central axis of the laser beam 24 is typically within ±30% of the beam diameter defined by D4σ.

[0059] In step S3, the additive manufacturing apparatus 100 forms a unit bead 51b in contact with the flattened unit bead 51a, thereby forming a bead layer. The unit bead 51b is formed at a position in contact with the flat portion 53 of the unit bead 51a.

[0060] In step S4, the additive manufacturing apparatus 100 determines whether the formation of the weld bead layer is complete. The object is a stacked body composed of multiple weld bead layers. If the formation of the weld bead layer is not complete (step S4, No), the additive manufacturing apparatus 100 proceeds to step S5. If the formation of the weld bead layer is complete (step S4, Yes), the additive manufacturing apparatus 100 proceeds to step S6.

[0061] After forming unit weld bead 51b in step S3, additive manufacturing apparatus 100 determines whether formation of the weld bead layer including unit weld beads 51a and 51b is complete. If formation of the weld bead layer is not complete at the time unit weld bead 51b is formed, additive manufacturing apparatus 100 irradiates edge 52 of formed unit weld bead 51b with laser beam 24 in step S5. Additive manufacturing apparatus 100 flattens unit weld bead 51b, thereby forming flat portion 53 at edge 52 of unit weld bead 51b.

[0062] After flattening unit bead 51b, additive manufacturing apparatus 100 returns to step S3. In step S3, additive manufacturing apparatus 100 forms unit bead 51c adjacent to flattened unit bead 51b. Additive manufacturing apparatus 100 repeats steps S3 through S5 until formation of one weld bead layer is complete.

[0063] Once the formation of the weld bead layer 54 is complete through the formation of the unit weld bead 51c, the additive manufacturing apparatus 100 proceeds to step S6. In step S6, the additive manufacturing apparatus 100 forms a weld bead layer on the formed weld bead layer 54. The additive manufacturing apparatus 100 forms the weld bead layer on the weld bead layer 54 through the same sequence as steps S1 to S5.

[0064] If the formation of the weld bead layer is completed, the additive manufacturing device 100 determines whether the formation of the object is completed in step S7. If the formation of the object is not completed (step S7, No), the additive manufacturing device 100 returns the sequence to step S6 and forms the weld bead layer. The additive manufacturing device 100 repeats the sequence of steps S6 and S7 to form each weld bead layer of the object. If the formation of the object is completed (step S7, Yes), the additive manufacturing device 100 ends. Figure 4 The actions involved are in the order shown.

[0065] Figure 7 This figure illustrates a comparative example of the first embodiment. The comparative example shows unit weld beads 51a, 51b, and 51c formed without flattening. Unit weld bead 51b is formed directly with the contact angle θa of unit weld bead 51a being large. This results in gaps 57 remaining between adjacent unit weld beads 51a and 51b and base material 17. Gaps 57 also remain between adjacent unit weld beads 51b and 51c and base material 17.

[0066] In the first embodiment, when unit weld beads forming a weld bead layer are formed in contact with each other, control device 1 flattens the formed unit weld beads by irradiating them with laser beam 24, thereby forming a unit weld bead in contact with the flattened unit weld bead. Control device 1 flattens the unit weld bead by irradiating edge 52 of the formed unit weld bead with laser beam 24.

[0067] After reducing the contact angle θa of the formed unit weld bead through flattening, the additive manufacturing apparatus 100 forms a unit weld bead in contact with the unit weld bead. The additive manufacturing apparatus 100 facilitates the flow of molten material 21 between adjacent unit weld beads and the workpiece, thereby suppressing the formation of voids. Consequently, the additive manufacturing apparatus 100 can suppress voids between adjacent unit weld beads within the same weld bead layer.

[0068] Furthermore, the object may include unit beads that have not been flattened and unit beads that are in contact with the unit beads. The additive manufacturing apparatus 100 may flatten the unit beads on portions of the object where strength improvement is desired, while omitting flattening on other portions.

[0069] The additive manufacturing apparatus 100 can perform a flattening process when the contact angle θa is greater than or equal to a threshold value, or when the flatness of a unit weld bead is less than a threshold value. A camera for observing the contact angle θa or the flatness can be mounted above the machining head 10. For example, the additive manufacturing apparatus 100 can perform a flattening process when the contact angle θa exceeds 45 degrees or when the flatness is less than 5. This reduces machining time compared to performing a flattening process on all unit weld beads and reduces the amount of voids remaining in the object.

[0070] Furthermore, when achieving a flat unit weld bead by fusing the molten material 21 instead of the flattening treatment described in Embodiment 1, increasing the output of the laser beam 24 or increasing the scanning speed are considered effective. To form a flat unit weld bead, it is necessary to improve the weldability of the molten material 21 to the surface 22. However, there are limitations on the conditions that can improve the weldability of the molten material 21 to the surface 22. Specifically, increasing the output of the laser beam 24 to achieve a flat unit weld bead will cause excessive melting of the wire 5, resulting in poor welding to the surface 22 and preventing the formation of a flat unit weld bead. Furthermore, excessively increasing the scanning speed will also result in poor welding and prevent the formation of a flat unit weld bead.

[0071] In the first embodiment, the formation of the unit bead and the flattening process can be processed separately, allowing the selection of optimal values ​​for the flattening process for parameters such as the output, scanning speed, and beam diameter of the laser beam 24. In the first embodiment, the flattening process is limited to the edge 52 of the unit bead. Therefore, the control device 1 can flatten the formed unit bead by irradiating the unit bead with a laser beam 24 having a lower intensity than that used during the unit bead formation. Alternatively, the control device 1 can flatten the formed unit bead by using a laser beam 24 with a higher scanning speed than that used during the unit bead formation. Thus, the additive manufacturing apparatus 100 can suppress excessive heat input into the workpiece and the unit bead, thereby minimizing the thermal effects on the workpiece and the unit bead. Consequently, the additive manufacturing apparatus 100 can achieve high-quality molding.

[0072] Here, good flattening means that the contact angle θa after the flattening process is less than or equal to 45 degrees, or the flatness is greater than or equal to 5. According to experiments conducted by the inventors, the ideal flatness is 7. If the additive manufacturing apparatus 100 can form a unit weld bead with a flatness greater than or equal to 5 even without performing the flattening process, the flattening process can be omitted.

[0073] When the temperature of the deposit 18 rises due to prolonged molding, the heat of the deposit 18 may flatten the unit weld beads. Therefore, flattening can be performed on weld bead layers where the temperature rise of the deposit 18 is minimal, such as the three weld bead layers stacked on the base material 17, while flattening can be omitted for each weld bead layer formed above these three weld bead layers. Alternatively, the intensity of the laser beam 24 during the flattening process can be reduced for each weld bead layer formed above these three weld bead layers. This allows the additive manufacturing apparatus 100 to prevent excessive heat input into the deposit 18. Furthermore, by omitting the flattening process, the additive manufacturing apparatus 100 can shorten processing time. This also applies to the second and third embodiments.

[0074] According to the first embodiment, when unit weld beads forming a weld bead layer are formed in contact with each other, additive manufacturing apparatus 100 flattens the formed unit weld beads by irradiating them with laser beam 24, thereby forming unit weld beads that contact the flattened unit weld beads. Additive manufacturing apparatus 100 can suppress the presence of gaps between adjacent unit weld beads in the same layer. As described above, additive manufacturing apparatus 100 is effective in suppressing a decrease in the strength of a product.

[0075] Implementation method 2.

[0076] In the second embodiment, Figure 1 Other additional processing methods implemented by the additional manufacturing device 100 shown in FIG. Figure 8 and Figure 9 The operation of the additive manufacturing apparatus 100 according to the second embodiment will be described.

[0077] Figure 8 This is a flowchart showing the operation procedure of the additive manufacturing apparatus according to the second embodiment. Figure 9 This figure illustrates the formation of a deposit achieved by an additive manufacturing apparatus according to a second embodiment. In the second embodiment, an additive manufacturing apparatus 100 performs a planarization process by irradiating the top of a unit weld bead with a laser beam 24. In the second embodiment, the same components as those in the first embodiment are denoted by the same reference numerals, and the description will focus on the components that differ from the first embodiment.

[0078] In step S1, the additive manufacturing apparatus 100 forms a unit weld bead 51a on the workpiece, i.e., the base material 17. In step S11, the additive manufacturing apparatus 100 irradiates the top 55 of the formed unit weld bead 51a with the laser beam 24. The additive manufacturing apparatus 100 scans the top 55 with the laser beam 24, thereby irradiating the entire top 55 with the laser beam 24. In the second embodiment, the entire unit weld bead 51a is flattened through a flattening process. The additive manufacturing apparatus 100 flattens the entire unit weld bead 51a, thereby reducing the contact angle θa of the unit weld bead 51a. The additive manufacturing apparatus 100 further irradiates the unit weld bead 51a with the laser beam 24 at the position irradiated by the laser beam 24 when the unit weld bead 51a was formed, thereby performing a flattening process.

[0079] Step S3 and step S4 are Figure 4 The same process is described for the first embodiment. In step S12, the additive manufacturing apparatus 100 irradiates the top portion 55 of the formed unit weld bead 51b with the laser beam 24. The additive manufacturing apparatus 100 flattens the entire unit weld bead 51b by performing a flattening process on the unit weld bead 51b. The additive manufacturing apparatus 100 repeats steps S3, S4, and S12 until one weld bead layer is formed. After the weld bead layer is formed, the additive manufacturing apparatus 100 proceeds to step S6.

[0080] If the formation of the weld bead layer is completed, the additive manufacturing device 100 is connected to the Figure 4 Similarly to the case of the embodiment 1 shown, each weld bead layer of the object is formed through steps S6 and S7. When the formation of the object is completed (step S7, Yes), the additive manufacturing apparatus 100 ends. Figure 8 The actions involved are in the order shown.

[0081] In the second embodiment, when unit weld beads forming a weld bead layer are formed in contact with each other, control device 1 flattens the formed unit weld beads by irradiating them with laser beam 24, thereby forming a unit weld bead in contact with the flattened unit weld bead. Control device 1 flattens the unit weld bead by irradiating top 55 of the formed unit weld bead with laser beam 24.

[0082] In the second embodiment, the additive manufacturing apparatus 100, similar to the first embodiment, can form a unit weld bead that contacts the unit weld bead after reducing the contact angle θa of the formed unit weld bead through flattening. In the second embodiment, the additive manufacturing apparatus 100 flattens the entire unit weld bead, thereby producing a unit weld bead with a smooth surface with minimal irregularities. The additive manufacturing apparatus 100 can suppress voids caused by the irregularities on the surface of the unit weld bead. Consequently, the additive manufacturing apparatus 100 can suppress voids remaining in the object.

[0083] In the second embodiment, the additive manufacturing apparatus 100 can irradiate a laser beam 24 having a higher intensity than when welding the molten material 21 during the flattening process. In this case, the control device 1 flattens the formed unit weld bead by irradiating the laser beam 24 having a higher intensity than when forming the unit weld bead. By irradiating the laser beam 24 having a higher intensity than when welding the molten material 21, the additive manufacturing apparatus 100 can obtain a highly flat unit weld bead. By obtaining a highly flat unit weld bead, the additive manufacturing apparatus 100 can further suppress voids.

[0084] In the second embodiment, the additive manufacturing apparatus 100 can irradiate a laser beam 24 having a larger diameter than that used when welding the molten material 21. In this case, the control device 1 flattens the formed unit weld bead by irradiating the laser beam 24 with a larger diameter than that used when forming the unit weld bead. By irradiating the laser beam 24 over a wide area, the additive manufacturing apparatus 100 can obtain highly flat unit weld beads. By obtaining highly flat unit weld beads, the additive manufacturing apparatus 100 can further suppress voids. The additive manufacturing apparatus 100 can vary the diameter of the laser beam 24 by driving the multiple lenses provided within the processing head 10 to achieve a zoom function.

[0085] Implementation method 3.

[0086] In embodiment 3, Figure 1 Other additional processing methods implemented by the additional manufacturing device 100 shown in FIG. Figures 10 to 14 , the operation of the additive manufacturing apparatus 100 according to the third embodiment will be described.

[0087] In the third embodiment, the additive manufacturing apparatus 100 forms a unit weld bead, ie, a beaded weld bead, by adding the molten material 21 while the machining head 10 is stopped.

[0088] Figure 10 This is a flowchart showing the operation procedure of the additive manufacturing apparatus according to the third embodiment. Figure 11 This is a diagram for explaining the formation of a deposit by the additive manufacturing apparatus according to the third embodiment. Figure 12 This is a flowchart illustrating a method for forming a beaded weld by the additive manufacturing apparatus according to the third embodiment. Figure 13 This is a diagram for explaining the formation of a beaded weld by the additive manufacturing apparatus according to the third embodiment. Figure 14This is a schematic plan view of a beaded weld bead formed by an additive manufacturing apparatus according to Embodiment 3. In Embodiment 3, additive manufacturing apparatus 100 performs a flattening process on the beaded weld bead. In Embodiment 3, components identical to those in Embodiments 1 or 2 are denoted by the same reference numerals, and the description will focus on components that differ from Embodiments 1 or 2.

[0089] Here, refer to Figure 12 and Figure 13 The formation of a beaded weld will be described below. In step S31, the additive manufacturing apparatus 100 moves the processing head 10 to align the center axis CL of the laser beam 24 with the center of the processing area 26. In step S32, the additive manufacturing apparatus 100 ejects the wire 5 obliquely toward the processing area 26, bringing the tip of the wire 5 into contact with the surface 22. Ejecting the wire 5 means advancing the wire 5 from the wire supply nozzle 12 to the irradiation position of the laser beam 24 toward the surface 22.

[0090] When the wire 5 is ejected from the wire supply nozzle 12 and contacts the surface 22, the center axis CW of the wire 5 and the center axis CL of the laser beam 24 intersect at the surface 22. Alternatively, the center axis CW of the wire 5 intersects the surface 22 within the diameter of the laser beam 24. Thus, the additive manufacturing apparatus 100 can form a beaded weld bead 61a centered at the intersection of the center axis CW of the wire 5 and the center axis CL of the laser beam 24 on the surface 22.

[0091] In step S33, the additive manufacturing apparatus 100 irradiates the processing area 26 with the laser beam 24. The laser beam 24 irradiates the wire 5 disposed in the processing area 26. In conjunction with the irradiation with the laser beam 24, the inert gas 25 is ejected from the gas nozzle 13 toward the processing area 26. Preferably, the inert gas 25 is ejected from the gas nozzle 13 for a predetermined period of time before irradiating the processing area 26 with the laser beam 24. This allows the additive manufacturing apparatus 100 to remove reactive gases, such as oxygen, remaining within the gas nozzle 13.

[0092] In step S34, the additive manufacturing apparatus 100 ejects the wire 5 from the wire supply nozzle 12 toward the surface 22, thereby starting to supply the wire 5 to the processing area 26. The molten material 21 produced by the irradiation of the laser beam 24 onto the wire 5 fuses with the surface 22. This forms a beaded weld 61a in the processing area 26. After starting to supply the wire 5 to the processing area 26, the additive manufacturing apparatus 100 continues to supply the wire 5 for a predetermined supply time.

[0093] The additive manufacturing apparatus 100 adjusts the rotational speed of the rotary motor 4, thereby adjusting the feed speed of the wire 5. The feed speed of the wire 5 is limited by the output of the laser beam 24. That is, there is a correlation between the feed speed of the wire 5 and the output of the laser beam 24, which is required to achieve proper welding of the molten material 21 to the processing area 26. By increasing the output of the laser beam 24, the time required to form the beaded weld 61a can be shortened. In addition, if the feed speed of the wire 5 is too fast relative to the output of the laser beam 24, the wire 5 does not melt and remains. If the feed speed of the wire 5 is too slow relative to the output of the laser beam 24, the wire 5 is excessively heated, causing the molten material 21 to fall from the wire 5 as droplets. In this case, the molten material 21 may be welded in a shape different from the desired shape.

[0094] The size of the beaded bead 61a can be adjusted by changing the supply time of the wire 5 and the irradiation time of the laser beam 24. By increasing the supply time of the wire 5 and the irradiation time of the laser beam 24, the diameter of the beaded bead 61a to be formed can be increased. On the other hand, by shortening the supply time of the wire 5 and the irradiation time of the laser beam 24, the diameter of the beaded bead 61a to be formed can be reduced.

[0095] After the beaded weld bead 61a is formed, the additive manufacturing apparatus 100 removes the wire 5 from the processing area 26 in step S35. In step S36, the additive manufacturing apparatus 100 stops irradiating the processing area 26 with the laser beam 24. Here, the gas nozzle 13 does not stop ejecting the inert gas 25 toward the workpiece, but continues ejecting it. Specifically, after the laser oscillator 2 stops, the gas nozzle 13 continues ejecting the inert gas 25 toward the processing area 26 for a predetermined duration.

[0096] The duration of the ejection of the inert gas 25 is the time required from the cessation of irradiation with the laser beam 24 until the temperature of the formed beaded bead 61a is reduced to a predetermined temperature by the inert gas 25. This duration is determined based on various conditions, such as the material of the wire 5 and the size of the beaded bead 61a. Information regarding the duration is pre-stored in the control device 1. After the laser beam 24 is stopped, the additive manufacturing apparatus 100 stops ejecting the inert gas 25 after the predetermined duration has elapsed. This completes the formation of one beaded bead 61a.

[0097] exist Figure 10 In step S21 shown in FIG. 1 , the additive manufacturing apparatus 100 forms a beaded weld bead 61a on the workpiece, that is, the base material 17, as described above. Figure 14In the beaded bead 61a shown, the width Wx in the X-axis direction and the width Wy in the Y-axis direction are the same. Widths Wx and Wy may also be different. The ratio of widths Wx to Wy, i.e., Wx / Wy, may be within the range of 0.5 to 2.0.

[0098] In step S22, the additive manufacturing apparatus 100 irradiates the formed bead 61a with the laser beam 24. In the third embodiment, the entire bead 61a is flattened through a flattening process. The additive manufacturing apparatus 100 flattens the entire bead 61a, thereby reducing the contact angle θa of the bead 61a. The additive manufacturing apparatus 100 performs the flattening process by re-irradiating the position where the laser beam 24 was irradiated when the bead 61a was formed with the laser beam 24.

[0099] In step S23, the additive manufacturing apparatus 100 forms a beaded weld bead 61b that is in contact with the flattened beaded weld bead 61a. In step S24, the additive manufacturing apparatus 100 determines whether formation of the weld bead layer is complete. If formation of the weld bead layer is not complete (step S24, No), the additive manufacturing apparatus 100 proceeds to step S25. If formation of the weld bead layer is complete (step S24, Yes), the additive manufacturing apparatus 100 proceeds to step S26.

[0100] After forming the beaded bead 61b in step S23, the additive manufacturing apparatus 100 determines whether formation of the weld bead layer, including the beaded beads 61a and 61b, is complete. If formation of the weld bead layer is not complete at the time of forming the beaded bead 61b, the additive manufacturing apparatus 100 irradiates the formed beaded bead 61b with the laser beam 24 in step S25. The additive manufacturing apparatus 100 repeats the sequence of steps S23 to S25 until formation of one weld bead layer is complete.

[0101] If the formation of the weld bead layer is completed, then in step S26, the additive manufacturing device 100 forms a weld bead layer on the formed weld bead layer. If the formation of the weld bead layer is completed, then in step S27, the additive manufacturing device 100 determines whether the formation of the object is completed. In the case that the formation of the object is not completed (step S27, No), the additive manufacturing device 100 returns the sequence to step S26 and forms the weld bead layer. The additive manufacturing device 100 repeats the sequence of steps S26 and S27, thereby forming each weld bead layer of the object. In the case that the formation of the object is completed (step S27, Yes), the additive manufacturing device 100 ends. Figure 10 The actions involved are in the order shown.

[0102] In the third embodiment, when beaded weld beads are formed that contact each other and constitute a weld bead layer, control device 1 flattens the formed beaded weld beads by irradiating laser beam 24, thereby forming a beaded weld bead that contacts the flattened beaded weld bead. Thus, additive manufacturing apparatus 100 can form a beaded weld bead that contacts the formed beaded weld bead after reducing the contact angle θa of the formed beaded weld bead through flattening.

[0103] In the third embodiment, during the flattening process, the additive manufacturing apparatus 100 irradiates a laser beam 24 having a higher intensity than that used when welding the molten material 21. In this case, the control device 1 flattens the formed bead by irradiating the laser beam 24 having a higher intensity than that used when forming the bead. By irradiating the laser beam 24 having a higher intensity than that used when welding the molten material 21, the additive manufacturing apparatus 100 can obtain a unit weld bead with high flatness. By obtaining a unit weld bead with high flatness, the additive manufacturing apparatus 100 can further suppress voids.

[0104] In the third embodiment, the additive manufacturing apparatus 100 can irradiate a laser beam 24 having a larger diameter than that used when welding the molten material 21. In this case, the control device 1 flattens the formed bead by irradiating the laser beam 24 with a larger diameter than that used when forming the bead. By irradiating the laser beam 24 over a wide area, the additive manufacturing apparatus 100 can obtain highly flat unit weld beads. By obtaining highly flat unit weld beads, the additive manufacturing apparatus 100 can further suppress voids.

[0105] In the third embodiment, the additive manufacturing apparatus 100 can flatten the bead by irradiating the edge of the bead with a laser beam 24. Alternatively, the additive manufacturing apparatus 100 can flatten the bead by irradiating the bead with a laser beam 24 having a lower intensity than the laser beam 24 used to form the bead. Furthermore, the additive manufacturing apparatus 100 can flatten the bead by irradiating the top of the bead with a laser beam 24. In either case, the additive manufacturing apparatus 100 can form a bead in contact with the bead after reducing the contact angle θa of the formed bead through the flattening process.

[0106] In the third embodiment, the additive manufacturing apparatus 100 stacks weld beads formed by juxtaposing bead-shaped beads to produce an object. Compared to a case where the unit weld beads are linear, the additive manufacturing apparatus 100 achieves improved shaping resolution, thereby improving shaping accuracy. This improved shaping resolution increases the number of interfaces where unit weld beads contact each other. This increase in interfaces increases the number of locations where voids can form. According to the third embodiment, the additive manufacturing apparatus 100 can achieve high-precision shaping while suppressing the presence of voids in the object.

[0107] Implementation method 4.

[0108] In the fourth embodiment, a machine learning device that learns the details of processing conditions for a planarization process will be described. Figure 15 This figure shows the configuration of an additive manufacturing system according to a fourth embodiment. The additive manufacturing system 200 according to the fourth embodiment includes an additive manufacturing device 100, a CAM (Computer Aided Manufacturing) device 110, and a machine learning device 120. The machine learning device 120 learns the details of the processing conditions used to flatten a unit weld bead having an ideal flatness. In the fourth embodiment, the same reference numerals are used for the same components as those in the first to third embodiments, and the description will focus on the components that differ from those in the first to third embodiments.

[0109] CAD (Computer-Aided Design) data is input to the CAM device 110. Based on the CAD data, the CAM device 110 generates a CAD model, design data that specifies the shape of the additively manufactured object. Based on the CAD model, the CAM device 110 generates a machining path for machining the object shape and creates a machining program based on the machining path data. The control device 1 controls the additive manufacturing apparatus 100 according to the machining program created by the CAM device 110.

[0110] The machine learning device 120 includes a state observation unit 71, a learning unit 72, and an operation result acquisition unit 75. The state observation unit 71 observes state information, including command values ​​generated during additive manufacturing and state quantities related to the processing state. The state observation unit 71 observes various command values ​​and state quantities related to the processing state generated by the control device 1 as state variables. These various command values ​​include the position command value output to the processing head drive device 14, the output command value output to the laser oscillator 2, and the supply command value output to the rotary motor 4. The state quantities include the temperature of the deposit 18.

[0111] The operation result acquisition unit 75 acquires shape information indicating the shape of the unit weld bead as an operation result. The shape information includes data on the width Wx, Wy, and height H of the unit weld bead. The shape information is acquired by measurement means such as a camera or height sensor 50 mounted above the machining head 10.

[0112] The learning unit 72 creates a data set by combining the state information input from the state observation unit 71 and the shape information input from the operation result acquisition unit 75. Based on the data set created based on the state information and shape information, the learning unit 72 learns the relationship between processing conditions and flatness. Processing conditions include conditions related to the material used for forming, the material of the base material 17, laser output, irradiation time, cooling time, and processing path.

[0113] The learning algorithm used by the learning unit 72 can use any learning algorithm. As an example, the case where reinforcement learning is applied is described. Reinforcement learning is an intelligent agent, i.e., an acting subject, in a certain environment observing the current state and deciding the action to be taken. The intelligent agent obtains rewards from the environment by selecting actions, and learns the countermeasures that obtain the most rewards after a series of actions. As representative methods of reinforcement learning, Q-learning and TD-learning are known. For example, in the case of Q-learning, the usual update formula of the action value function Q(s, a), i.e., the action value table, is expressed by the following formula (1). The action value function Q(s, a) represents the value of the action of selecting action "a" based on the environment "s", i.e., the action value Q.

[0114] [Formula 1]

[0115] Q(s t , a t )←Q(s t , a t )+α(r t+1 +γmax a Q(s t+1 , a t )-Q(s t , a t )) …(1)

[0116] The update formula expressed by equation (1) above indicates that if the action value of the best action "a" at time "t+1" is greater than the action value Q of action "a" performed at time "t," action value Q is increased; otherwise, action value Q is decreased. In other words, the action value function Q(s, a) is updated so that the action value Q of action "a" at time "t" approaches the best action value at time "t+1." Thus, the best action value in a given environment is propagated to the action value in the previous environment.

[0117] The learning unit 72 includes a reward calculation unit 73 and a function update unit 74. The reward calculation unit 73 calculates rewards based on the state information and shape information. The function update unit 74 updates the function used to determine the machining conditions, that is, the relationship between the conditions, according to the rewards calculated by the reward calculation unit 73.

[0118] The reward calculation unit 73 calculates the reward "r" based on the difference between the flatness of the unit weld bead and the ideal value representing the ideal flatness. For example, if the result of changing one of the processing conditions is that the difference between the flatness of the unit weld bead and the ideal value is less than or equal to a threshold value, the reward calculation unit 73 increases the reward "r". The reward calculation unit 73 increases the reward "r" by assigning a reward value of "1". The reward value is not limited to "1". In addition, if the result of changing one of the processing conditions is that the difference between the flatness of the unit weld bead and the ideal value is greater than a threshold value, the reward calculation unit 73 decreases the reward "r". The reward calculation unit 73 decreases the reward "r" by assigning a reward value of "-1". The reward value is not limited to "-1".

[0119] Figure 16 This is a flowchart showing the operation procedure of the machine learning device according to the fourth embodiment. Figure 16 Flowchart of , illustrating the reinforcement learning method for updating the action-value function Q(s, a).

[0120] In step S41, the machine learning device 120 obtains data on the height and width of the flattened unit weld bead. In step S42, the machine learning device 120 calculates the flatness of the unit weld bead based on the data on the height and width of the unit weld bead. In step S43, the machine learning device 120 calculates the difference between the calculated flatness and the ideal value. The ideal value is set to 7, for example.

[0121] In step S44, the machine learning device 120 calculates the reward based on the difference. In step S45, the machine learning device 120 updates the action-value function Q(s, a) based on the reward. In step S46, the machine learning device 120 determines whether the action-value function Q(s, a) has converged. The machine learning device 120 determines that the action-value function Q(s, a) has converged because the update of the action-value function Q(s, a) in step S45 cannot be performed.

[0122] If it is determined that the action value function Q(s, a) has not converged (step S46, No), the machine learning device 120 returns the operation sequence to step S41. If it is determined that the action value function Q(s, a) has converged (step S46, Yes), the machine learning device 120 ends the learning of the learning unit 72. Thus, the machine learning device 120 ends Figure 16 The machine learning device 120 may not perform the determination in step S46 but may continue the learning by returning the operation sequence from step S45 to step S41.

[0123] The machine learning device 120 stores the generated action-value function Q(s, a) as a trained model. Based on the trained model, the control device 1 infers the machining conditions for forming a unit weld bead with ideal flatness and adjusts the machining conditions based on the inference result.

[0124] In the fourth embodiment, reinforcement learning is described as being applied to the learning algorithm used by the learning unit 72. However, learning algorithms other than reinforcement learning may also be applied. The learning unit 72 may perform machine learning using a well-known learning algorithm other than reinforcement learning, such as deep learning, neural networks, genetic programming, functional logic programming, or support vector machines.

[0125] The machine learning device 120 is not limited to being included in the additive manufacturing system 200 and may also be a device external to the additive manufacturing system 200. The machine learning device 120 may be a device that can be connected to the additive manufacturing system 200 via a network. The machine learning device 120 may also be a device that exists on a cloud server. The machine learning device 120 may also be built into the control device 1.

[0126] The learning unit 72 can learn the relationship between processing conditions based on the data set created for multiple additive manufacturing devices 100. The learning unit 72 can obtain the data set from multiple additive manufacturing devices 100 used at the same site, or can also obtain the data set from multiple additive manufacturing devices 100 used at different sites. The data set can be data collected from multiple additive manufacturing devices 100 operating independently of each other at multiple sites. After the collection of data sets from multiple additive manufacturing devices 100 is started, a new additive manufacturing device 100 can be added to the objects from which the data sets are collected. In addition, after the collection of data sets from multiple additive manufacturing devices 100 is started, some of the multiple additive manufacturing devices 100 can be excluded from the objects from which the data sets are collected.

[0127] The learning unit 72 that has performed learning on a particular additive manufacturing device 100 can also perform learning on other additive manufacturing devices 100 other than that device 100. The learning unit 72 that has performed learning on the other additive manufacturing device 100 can update the output prediction model by re-learning on the other additive manufacturing device 100. The machine learning function or the trained model resulting from the learning in Embodiment 4 can be incorporated into CAM software for additive manufacturing.

[0128] According to the fourth embodiment, the machine learning device 120 outputs a trained model for forming unit weld beads with ideal flatness to the control device 1. The additive manufacturing apparatus 100 adjusts processing conditions based on the trained model, thereby forming unit weld beads with ideal flatness. This allows the additive manufacturing apparatus 100 to suppress voids remaining in the object.

[0129] The structures shown in the above embodiments illustrate an example of the content of the present invention. The structures of the embodiments can be combined with other known technologies. The structures of the embodiments can also be appropriately combined with each other. A part of the structure of the embodiments can be omitted or changed without departing from the scope of the present invention.

[0130] Description of the label

[0131] 1 Control unit, 2 Laser oscillator, 3 Optical cable, 4 Rotating motor, 5 Wire, 6 Wire reel, 7 Gas supply device, 8 Piping, 10 Processing head, 11 Beam nozzle, 12 Wire supply nozzle, 13 Gas nozzle, 14 Processing head drive, 15 Work table, 16 Rotating mechanism, 17 Base material, 18 Deposition, 19 Material supply unit, 21 Molten material, 22 Surface, 23 Molten pool, 24 Laser beam, 25 Inert gas, 26 Processing area, 41 CPU, 42 RAM, 43 ROM, 44 storage device, 45 input / output interface, 46 bus, 50 height sensor, 51a, 51b, 51c unit weld beads, 52 edge, 53 flat portion, 54 weld bead layer, 55 top, 57 gap, 61a, 61b bead weld beads, 71 state observation unit, 72 learning unit, 73 feedback calculation unit, 74 function update unit, 75 action result acquisition unit, 100 additive manufacturing device, 110 CAM device, 120 machine learning device, 200 additive manufacturing system.

Claims

1. An additive manufacturing apparatus for manufacturing a shaped object by stacking layers of unit weld beads, which are solidified products of a molten material, in parallel. The additive manufacturing device is characterized by having: a material supply unit that supplies the material to a workpiece; an irradiation portion that irradiates a light beam to melt the supplied material; and a control device that forms the unit weld bead by controlling the material supply unit and the irradiation unit, When the unit weld beads forming the layer are in contact with each other, the control device flattens the formed unit weld beads by irradiating the light beam, and forms a unit weld bead in contact with the flattened unit weld beads. The additive manufacturing device has a camera for observing the contact angle or flatness of the unit weld bead. The control device flattens the unit weld bead when the contact angle of the unit weld bead exceeds 45 degrees or the flatness is less than 5.

2. The additive manufacturing device according to claim 1, characterized in that The control device flattens the formed unit weld bead by irradiating the edge of the formed unit weld bead with the light beam.

3. The additive manufacturing device according to claim 2, characterized in that The control device flattens the formed unit bead by irradiating the light beam having a lower intensity than the light beam used when forming the unit bead.

4. The additive manufacturing device according to claim 1, characterized in that The control device flattens the formed unit weld bead by irradiating the top of the formed unit weld bead with the light beam.

5. The additive manufacturing device according to claim 4, characterized in that The control device flattens the formed unit weld bead by irradiating the light beam having a higher intensity than the light beam used when forming the unit weld bead.

6. The additive manufacturing device according to claim 4, characterized in that The control device flattens the formed unit weld bead by irradiating the light beam having a larger diameter than the light beam used when forming the unit weld bead.

7. The additive manufacturing device according to any one of claims 1 to 6, characterized in that The unit weld bead is a bead-shaped weld bead.

8. The additive manufacturing device according to claim 2, wherein: The control device flattens the formed unit bead using the light beam at a higher scanning speed than that of the light beam when forming the unit bead.

9. The additive manufacturing device according to any one of claims 1 to 6, characterized in that The control device forms the layer including the plurality of flattened unit beads.

10. The additive manufacturing device according to any one of claims 1 to 6, characterized in that The material is thread.

11. An additive manufacturing method, wherein an additive manufacturing apparatus manufactures a shaped object by laminating layers of solidified molten material, namely, unit weld beads, in parallel. The additive manufacturing method is characterized by comprising the following steps: supplying the material to a workpiece; irradiating a light beam to melt the supplied material; and The unit weld bead is formed by supplying the material to the workpiece and controlling the irradiation of the light beam. When the unit weld beads forming the layer are in contact with each other, the formed unit weld beads are flattened by irradiation with the light beam, and a unit weld bead in contact with the flattened unit weld bead is formed. The additive manufacturing device has a camera for observing the contact angle or flatness of the unit weld bead. When the contact angle of the unit weld bead exceeds 45 degrees or the flatness is less than 5, the unit weld bead is flattened.

12. A machine learning device for learning the contents of processing conditions for additive manufacturing performed by an additive manufacturing apparatus, wherein when unit weld beads constituting layers are formed by solidifying molten material and contacting each other, the unit weld beads are flattened by irradiation with a light beam, a unit weld bead is formed that contacts the flattened unit weld beads, and the layers are stacked to produce an object. The machine learning device is characterized by having: a state observation unit that observes state information including a command value generated during the additive manufacturing and a state quantity related to a machining state; an operation result acquisition unit that acquires shape information indicating the shape of the flattened unit weld bead as an operation result; and a learning unit that learns the processing conditions for forming the flattened unit weld bead according to a data set created based on the state information and the shape information, The additive manufacturing device has a camera for observing the contact angle or flatness of the unit weld bead. When the contact angle of the unit weld bead exceeds 45 degrees or the flatness is less than 5, the unit weld bead is flattened.

Citation Information

Patent Citations

  • Laser processing apparatus and machine learning device

    CN108568596A

  • Method for manufacturing laminated molding and laminated molding

    JP2019063858A